Alphaviruses comprise a group of genetically, structurally, and serologically related arthropod-borne viruses of the Togaviridae family. These viruses are distributed worldwide, and persist in nature through a mosquito to vertebrate cycle. Birds, rodents, horses, primates, and humans are among the defined alphavirus vertebrate reservoir/hosts.
Twenty-six known viruses and virus subtypes have been classified within the alphavirus genus utilizing the hemagglutination inhibition (HI) assay. This assay segregates the 26 alphaviruses into three major complexes: the Venezuelan equine encephalitis (VEE) complex, the Semliki Forest (SF) complex, and the western equine encephalitis (WEE) complex. In addition, four other viruses, eastern equine encephalitis (EEE), Barmah Forest, Middelburg, and Ndumu, receive individual classification based on the HI serological assay.
Members of the alphavirus genus also are classified based on their relative clinical features in humans: alphaviruses associated primarily with encephalitis, and alphaviruses associated primarily with fever, rash, and polyarthritis. Included in the former group are the VEE and WEE complexes, and EEE. In general, infection with this group can result in permanent sequelae, including death. In the latter group is the SF complex, comprised of the individual alphaviruses Semliki Forest, Sindbis, Ross River, Chikungunya, O'nyong-nyong, and Mayaro. Although serious epidemics have been reported, infection by viruses of this group is generally self-limiting, without permanent sequelae.
Sindbis virus is the prototype member of the Alphavirus genus of the Togaviridae family. Its replication strategy is well characterized and serves as a model for other alphaviruses (Strauss and Strauss, 1994, Microbio. Rev., 58:491-562). The genome of Sindbis virus (like other alphaviruses) is an approximately 12 kb single-stranded, positive-sense RNA molecule that is capped and polyadenylated. Genome RNA is contained within a virus-encoded capsid protein shell which is, in turn, surrounded by a host-derived lipid envelope from which two viral-specific glycoproteins, E1 and E2, protrude as spikes from the virion surface. Certain alphaviruses (e.g., SF) also maintain an additional protein, E3, which is a cleavage product from the E2 precursor protein, PE2.
After virus particle absorption to target cells, penetration, and uncoating of the nucleocapsid to release viral genomic RNA into the cytoplasm, the replication process is initiated by translation of four nonstructural replicase proteins (nsP1-nsP4) from the 5' two-thirds of the viral genome. The four nsPs are translated as one of two polyproteins (nsP123 or nsP1234), and processed post-translationally into mature monomeric proteins by an active protease in the C-terminal domain of nsP2. Both of the nonstructural polyproteins and their derived monomeric units may participate in the RNA replication process, which involves nsP binding to the conserved nucleotide sequence elements (CSEs) present at the 5' and 3' ends, and an internal subgenomic junction region promoter.
The positive strand genome RNA serves as template for the nsP-catalyzed synthesis of a full-length complementary negative strand RNA. Synthesis of the negative strand RNA is catalyzed by binding of an nsP complex to the 3' terminal CSE of the positive strand genome RNA. The negative strand, in turn, serves as template for the synthesis of additional positive strand genome RNA, as well as an abundant subgenomic RNA, initiated internally at the junction region promoter. Synthesis of additional positive strand genome RNA occurs after binding of an nsP complex to the 3' terminal CSE of the complementary negative strand genome-length RNA template. Synthesis of the subgenomic mRNA from the negative strand RNA template is initiated from the junction region promoter. Thus, the 5' end and junction region CSEs of the positive strand genome RNA are functional only after being transcribed into the negative strand RNA complement (i.e., the 5' end CSE is functional when it is the 3' end of the genomic negative stranded complement).
Alphavirus structural proteins (sPs) are translated from the subgenomic RNA, which represents the 3' one-third of the genome, and like the nsPs, are processed post-translationally into the individual proteins (FIG. 1). Translation of this subgenomic mRNA produces a single polyprotein consisting of the structural proteins capsid (C) glycoprotein E2 and glycoprotein E1, plus the corresponding leader/signal sequences (E3, 6k) for glycoprotein insertion into the endoplasmic reticulum. The structural gene polyprotein is processed into the mature protein species by a combination of viral (capsid autoprotease) and cellular proteases (e.g., signal peptidase). Alphavirus structural proteins are produced at very high levels due to the abundance of subgenomic mRNA transcribed, as well as the presence of a translational enhancer element (Frolov and Schlesinger, 1994, J. Virol. 68:8111-8117; Sjoberg et al., 1994, Bio/Technol. 12:1127-1131) within the mRNA, located in the capsid gene coding sequence. Because all structural proteins are synthesized at equimolar ratios, as part of the polyprotein, the translation enhancer element exerts its effect equally on each of the genes.
The general strategy for construction of alphavirus-based expression vectors has been to substitute the viral structural protein genes with a heterologous gene, maintaining transcriptional control via the highly active subgenomic RNA promoter. Vectors of this configuration are termed RNA "replicons" and may be transcribed in vitro from cDNA using a bacteriophage promoter, or, generated in vivo directly from DNA when linked to a eukaryotic promoter. Currently, alphavirus replicon RNA is packaged into recombinant vector particles by transient co-transfection with in vitro transcribed defective helper RNA, or, using stable packaging cell lines having structural protein expression cassettes. The structural protein expression cassette(s) used for vector packaging encode either the intact "native" alphavirus structural polyprotein that is post-translationally processed into mature C, E2, and E1; or, alphavirus structural proteins that have been split into separate cassettes encoding either C or E2/E1.
As described in the detailed description and examples below, the present invention provides new compositions and methods for packaging of alphavirus particles, including novel alphavirus structural polyprotein genes, expression cassettes containing the genes, polypeptides expressed from the cassettes, and methods of using the genes, cassettes, and polypeptides for the high level packaging of alphavirus vector RNA into recombinant alphavirus vector particles in the absence of contaminating replication-competent virus. The invention also provides modifications to alphavirus vector systems wherein expression of the heterologous transgene from an alphavirus vector replicon may be reduced (suppressed) in desired cells, including during the vector packaging process. This specific reduction in transgene expression can provide a mechanism to increase the level of recombinant alphavirus vector particles produced in those instances where the transgene has a negative effect on replicon particle titer.